Re: TSVDIR review of draft-ietf-intarea-shared-addressing-issues-02

[Joe Touch <[e-mail]>]

On 2/2/2011 5:30 PM, Fernando Gont wrote:
> On 02/02/2011 10:24 p.m., Fernando Gont wrote:
>>> On 2/2/2011 5:04 PM, Fernando Gont wrote:
>>> ...
>>>>> At the least, it's worth noting that geolocation is already broken by
>>>>> tunnels, and that IP addressing does not ensure geographic proximity
>>>>> before attributing breakage on NATs or other sharing.
>>>>
>>>> Tunnels need not break geo-location. -- They do not masquerade the
>>>> source address. Or am I missing something?
>>>
>>> When I tunnel using an ISI address, whomever sees my address thinks I'm
>>> in California.
>> [..]
>
> And one might argue that, in this type of scenario, this "breakeage" of
> geo-location might, in some cases, be desirable.

It can't break if it was never intended to work.

Anyone who thinks that IP address = location is as incorrect as those 
who assume area code = location these days.

I.e., the overall assumption was never true. The fact is that it 
*sometimes* works, that's all.

Joe